Nonperforming Debts in Chinese
Enterprises: Patterns, Causes, and
Implications for Banking Reform
*Geng Xiao
The University of Hong Kong Hong Kong, SAR, China [email protected]
1 LINE SHORT REGULAR 1 LINE LONG Abstract
Given the domination of bank financing, nonperforming debts (NPDs) in large Chinese enterprises are a proxy for nonperform-ing loans (NPLs) in China’s major banks. Usnonperform-ing a firm-level survey of more than 20,000 large and medium-sized industrial enter-prises conducted by the National Bureau of Statistics of China, this paper estimates both the level and ratio of NPDs across ownership type, industry, and region for the period 1995–2002. The results show that NPD ratios have been falling since 2000 as a result of the rapid expansion of better-performing non-state terprises (NSEs), the improved performance of state-owned en-terprises (SOEs), and the exit of poor-performing enen-terprises (which has been facilitated by asset management companies and other merger and acquisition activities). SOEs, however, are still much more likely than NSEs to generate NPDs. This paper pro-vides useful tools and sector information for assessing enterprise debt risks and draws lessons for banking reform in China.
1. Introduction
Since early 2004, the newly established China Banking Regulatory Commission (CBRC) has announced quarterly
*This paper was presented at the Asian Economic Panel meeting in Hong Kong on 12–13 April 2004, and at the Hong Kong Mon-etary Authority workshop on 8 July 2004. The author would like to acknowledge research collaboration with the Industry and Transportation Department of the National Bureau of Statistics (NBS) of China. Xing Junling and Yu Xiaoyun provided exten-sive technical support at NBS. Tu Zhengge provided excellent research assistance. The author would like to thank the Research Grant Council of Hong Kong and the University Grant Council of Hong Kong for ªnancial support (project nos. HKU7167/98H and AOE/H05/99). This research work also beneªted from a
statistics on the ratio of nonperforming loans (NPLs) for China’s major banking in-stitutions. It reported a sharp decline in the NPL ratio from above 24 percent in 2002 to 19.6 percent in June 2003, 17.8 percent in December 2003, 13.3 percent in June and December 2004, and only 8.7 percent in June 2005. During the 2-year period between June 2003 and June 2005, the outstanding amount of NPLs in China dropped from RMB2.538 trillion to RMB1.276 trillion, a decrease of RMB1.262 trillion. This is clearly a dramatic turnaround in terms of banking sector performance. Are the ofªcial statistics reliable? What happened to the quality of bank loans and debt in China’s enterprises? These questions motivate this paper. Outside observers inter-preted the ofªcial reports cautiously because they did not understand how China’s banks calculated their NPL ratios. Most analysts and commentators would still esti-mate China’s NPL ratio to be at a much higher level than the ofªcial ªgure—usually two to three times that of the ofªcial NPL ratio. For example, UBS, an investment bank and global asset management ªrm, (Anderson 2005) estimates that China’s NPLs fell from about 50–55 percent in 1997–98 to about 25–30 percent by the end of 2004. These market estimations of NPL ratios are usually based on macroeconomic data because it is difªcult to get reliable and representative microeconomic data from Chinese banks.
This paper attempts to develop an alternative approach to the study of NPLs in China using ªrm-level microeconomic data. As a result of the limited development of stock markets and enterprise bond markets in China, banks are still the major holders of enterprises’ long-term and short-term debt. In recent years, Chinese banks have expanded rapidly in the business of consumer loans, especially mort-gage loans. The outstanding amounts of consumer loans rose from below 1 percent in 1998 to above 10 percent in 2004. Because the quality of consumer loans is gener-ally much better than that of enterprise loans, the quality of bank loans depends largely on the quality of the bank lending to enterprises. The quality of enterprise debt is directly linked to the proªtability of the enterprises. The ability to pay the in-terest and principal of loans derives ultimately from proªtability and cash income ºows of the enterprises. This is especially true if we are examining a large group of enterprises, in which the variations in the enterprise-speciªc timing of cash income ºows and the structures of ªnancing within the group will be averaged out statisti-cally through the law of large numbers, making the proªtability of each enterprise
1 LINE SHORT REGULAR 1 LINE LONG
seed fund for strategic research in corporate and ªnancial law and policy from The University of Hong Kong. The author is also grateful to Wing Thye Woo, Dwight Perkins, Jeffrey Sachs, Alan Siu, Richard Wong, Sung Yun Wing, Zhang Weiying, Erh-Cheng Hua, Ren Ruoen, Peng Wensheng, and Guonan Ma for comments and suggestions. The author takes responsibility for any errors.
the single most important contribution to the quality of the enterprise group’s port-folio of debts.
This paper uses the proªtability conditions of each enterprise to measure and char-acterize the quality of the enterprise group’s portfolio of debts. It uses both reported proªtability and imputed proªtability (the latter is derived from the components of value-added) to give two alternative estimates of the quality of debt portfolios for different groups of enterprises classiªed by ownership, industry, and region. The method is applied to a comprehensive annual survey of all the large and medium-sized industrial enterprises in China conducted by the National Bureau of Statistics of China. Information about the survey data can be found in tables A.1 to A.6 in the appendix. The survey sample includes more than 20,000 enterprises and covers the period from 1995 to 2002 (see table A.6). In 2002, the sample enterprises had 26 mil-lion employees in total, which is about 16.7 percent of China’s total industrial em-ployment. These enterprises incurred RMB5.7 trillion in debt, an amount equal to 43.6 percent of the total loans in China’s ªnancial institutions. The sample enter-prises contributed about 19.2 percent of China’s GDP. Clearly these enterenter-prises are the most important leaders in the Chinese industrial sector. Aggregate ªnancial in-formation about the sample enterprises has been regularly reported in the Statistical
Yearbookof China.
The contribution of this paper is the methodology of using the disaggregated ªrm-level data to study the enterprise proªtability of the enterprises and the quality of their debts. The paper derives both the level and ratio of NPD across enterprise groups by ownership, industry, and region for the period 1995–2002. The results show that NPDs have indeed been falling as a result of both the rapid expansion of better-performing non-state enterprises (NSEs) and improvements in the perfor-mance of state-owned enterprises (SOEs), as well as rapid exit of poor-performing state-owned enterprises, which has been facilitated by newly established asset man-agement companies that specialize in dealing with NPLs. The micro-level evidence uncovered here is largely consistent with the CBRC reports on falling NPLs and NPL ratios in China’s banking sector. This study, however, provides a more trans-parent, simpler, and more objective method of estimating NPDs and allows outsid-ers to examine the detailed causes and dynamics of the changing patterns of NPDs in Chinese enterprises. In particular, it was found that the SOEs had consistently generated higher NPD ratios than the NSEs, providing a challenge as well as an op-portunity for future banking reform.
Section 2 deªnes our concepts of NPDs; section 3 shows the patterns of NPD and NPD ratios across enterprises’ ownership type, industry, and region; and section 4
1 LINE SHORT REGULAR 1 LINE LONG
examines the trend of NPD ratios during the period 1995–2002 and provides scenar-ios of the future of NPD ratscenar-ios in China. Section 5 uses panel data regressions to identify the impacts of various factors on the proªtability and debt quality of the en-terprises. Section 6 addresses the sample-selection bias in the measure of NPD ratios resulting from the exit of poor-performing enterprises from the sample over the study period. Section 7 discusses the implications of the empirical results for bank-ing reform in China.
2. Defining and estimating NPDs
In recent years, the CBRC has been trying hard to monitor and supervise the NPLs in China’s banking institutions. It developed detailed rules on the reporting of the amount of NPLs and the NPL ratios. The purpose is to manage and reduce ªnancial risk by monitoring both the changing distribution of NPLs and the changing NPL ratios of individual banks and bank branches. This is clearly necessary and useful. Poor governance of banks is a sufªcient condition for creating NPLs, even when the enterprise sector is doing well.
The efforts of the CBRC and the individual banks to reduce NPLs, however, are only necessary conditions. Ultimately, the quality of China’s banking assets and enter-prise debts depends directly on the proªtability of China’s enterenter-prises. For example, in the short run it is easy for banks to reduce NPL ratios, or even the amount of NPLs, by simply expanding the total amount of loans. New loans are much less likely to have repayment problems in the short run, but they could create more bad loans in the future if they are extended to potentially loss-making enterprises. New loans can help existing loss-making enterprises continue to pay their interest on old loans, thereby also shifting the underlying risks to the future. The problem is espe-cially serious when the economy is booming. These are the main reasons why the re-liability of NPL statistics as reported by China’s banks can vary considerably, de-pending on how they are calculated. Outsiders are unable to completely understand how the NPLs and NPL ratios in China’s banks are actually calculated because the decisions in each individual case require judgments that are too complex for outsid-ers to assess. This is why analysts and commentators rely more on the study of China’s macroeconomic conditions, such as business cycles and sector performance, to gauge the level of NPLs in China. Based on their personal impressions and un-derstandings about the Chinese economy, they report NPL ratios that are usually two to three times larger than the ofªcial ratio.
This paper attempts to develop an objective measure of the quality of enterprise debts in China. Proªtability is the only criterion used in measuring the quality of these debts. The concept and its implementation are straightforward. If the enter-1 LINE SHORT
REGULAR 1 LINE LONG
prises are making proªts, the quality of their debts (more speciªcally their total lia-bilities) are termed “performing.” If they are making losses, their debts are regarded as “nonperforming.” The amount of NPD for a speciªc enterprise group is then the sum of the total liabilities in the loss-making enterprises for that group. The NPD ra-tio for the group is simply the rara-tio of the sum of total liabilities in the loss-making enterprises divided by the sum of total liabilities in both the loss-making and proªt-making enterprises in the group. These simple deªnitions of NPD and the NPD ra-tio make our NPD statistics objective, easy to measure, and easy to understand. Our concept of the NPD ratio, however, is not applicable to an individual enterprise, because according to our deªnition, an enterprise cannot have part of its total debt be performing and the other part be nonperforming: all the debts in any given enter-prise have the same quality. For instance, one enterenter-prise cannot have 70 percent of its debts performing and 30 percent nonperforming. For an individual enterprise, losses might occur in the ªrst few years, but the ªrm would make proªts in the fu-ture. After a close examination by its creditors, such a ªrm would be considered to have high-quality debt. Our deªnition of NPD would not be fair to this particular enterprise. On the other hand, a currently proªtable enterprise might become a loss maker. Then its debt quality would be bad upon close examination. Our assessment, based only on current proªtability, might not do justice to this particular enterprise. The difference between current proªtability and longer-term proªtability, however, could be seen as a random distribution for a large group of enterprises, such as groups in our sample separated by time, ownership, industry, and region. With a sizable group, the variability in the timing of cash ºows, proªt streams, payments to creditors, and other proªtability-related variables for enterprises within the group would offset each other, leaving the average NPD ratio for the group a much more reliable and accurate measure of the quality of the group’s portfolio of debts. This is why our concept of the NPD ratio is useful only for measuring debt quality for groups of enterprises. This study would be very useful as complementary research to the traditional method of estimating NPLs.
For our method to be useful, it needs to be applied to a representative sample with sizable groups of enterprises. The annual survey of large and medium-sized indus-trial enterprises conducted by China’s National Bureau of Statistics (NBS) is a suit-able data set for our method. The NBS data are in fact census data, not really sample data, because the survey covers all large and medium-sized industrial enterprises in China. In 2002, China had more than 180,000 industrial enterprises with sales above RMB5 million. The NBS sample includes only about 21,000–23,000 large and me-dium-sized industrial enterprises out of a total of 180,000 ªrms. Many small indus-trial enterprises are not included in our study, but most of them have limited access to bank ªnance under the current ªnancial system in China.
1 LINE SHORT REGULAR 1 LINE LONG
One major weakness of using our method for the NBS data is sample selection bias. The group of enterprises included in the NBS survey each year is not always the same. About 20 percent of the group consists of enterprises that enter and exit the survey sample each year as a result of changes in their size classiªcation or organi-zational changes such as mergers and acquisition, privatization, reorganization, and bankruptcy. The proªtability of exiting ªrms is not necessarily the same as that of the new entries. This means that we are studying only the largest and most dynamic frontier industrial enterprises in China and leaving out the poor-performing ones. We will address this issue in section 6 and show the impact of this sample selection bias on the estimated ratio of NPD.
Our sample covers not only SOEs but also other types of enterprises in all industries and regions, including rural and urban collectives, private enterprises, domestic mixed-ownership corporations, foreign-invested enterprises, and enterprises with investment from Hong Kong, Macau, and Taiwan. Our sample does not include nonindustrial enterprises, however, such as China Telecom, a big service sector ªrm. It is entirely possible that the enterprises not included in our sample have worse performance records than the enterprises in the NBS sample. If that is the case, the NPD ratios for the entire Chinese enterprise sector would be higher than the ratios reported in this study. Also, if the banking institutions in China are performing worse than the enterprises in the NBS sample, as a consequence of their own weak governance, the overall NPL situation of the Chinese economy as a whole would be worse than that for the sample enterprises used in this study. In addition, we cannot directly compare the NPL ratios reported by the CBRC with the NPD ratios esti-mated in our study because they are deªned differently. The NPD ratios here are de-signed to examine the trend and the cross-sectional patterns of the quality of Chi-nese enterprise debts.
During the period from 1995 to 2002, the sample enterprises created about 16–25 percent of China’s industrial employment and 33–43 percent of China’s industrial value added (see table A.6). Most signiªcantly, the sample enterprises contributed about 14–19 percent of China’s GDP. Their total liabilities, one of the key variables we examine in this paper, amount to about 43–65 percent of China’s total banking loans during the 1995–2002 period. Of course, not all of the total liabilities in the sample enterprises correspond to loans from banks. But even assuming that 60 per-cent of the total liabilities in the sample are related to various bank loans, the statisti-cal analysis in the paper provides an in-depth study of the quality of about 27–29 percent of China’s total loans. In summary, although the members of the sample en-terprises group are changing each year, as a whole they form a stable club of China’s elite industrial enterprises. The performance of this elite group of enterprises is 1 LINE SHORT
REGULAR 1 LINE LONG
much more representative of the performance of China’s industrial economy than, for example, the performance of China’s listed companies or any small sample of Chinese enterprises. Given the growing importance of China’s industrial sector for both the domestic and global economy, our analysis in this paper ªlls a crucial gap in understanding the dynamics of China’s industrial reform and development. Tables 1, 2, and 3 show the distribution of the total liabilities for the sample enter-prises by ownership, industry, and region. The objective is to ªnd out how much of these debts are located in proªt-making and loss-making enterprises and then calcu-late the amounts and ratios of the NPD. There are two underlying forces affecting the NPD ratios: the shifting distribution of debts across enterprise groups with dif-ferent proªtability and the changing proªtability of each group.
Table 1 shows the distribution of total liabilities (or total debts) across ownership types. The share of total debts by SOEs fell sharply, from 76.4 percent in 1995 to 48.2 percent in 2002, to the beneªt of domestic mixed-ownership corporations and pri-vate enterprises. The total debt for SOEs increased from RMB2.5 trillion in 1995 to RMB3.2 trillion in 1998, immediately before the Asian ªnancial crisis, and then fell to RMB2.8 trillion in 2002. The total debt for collectives followed the same pattern as that for SOEs, rising from RMB227 billion in 1995 to RMB287 billion in 1998 and then falling to RMB219 billion in 2002. The shifting of debt toward private, mixed, foreign, or overseas Chinese enterprises was steady and rapid throughout the 1995– 2002 period, without any interruption resulting from the Asian ªnancial crisis in 1998–99. For the 8 years from 1995 to 2002, total debt in the sample enterprises in-creased by RMB2,436 billion. Of this net increase, RMB246 billion ended up in the SOEs, RMB1,411 billion went to mixed-ownership enterprises, RMB467 billion to foreign enterprises, and RMB95 billion to private enterprises. The drastic changes in the distribution of total debt are strong evidence of the rapid but quiet privatization and opening up of the most dynamic part of China’s industrial sector. In the next section, we will show that the redistribution of total debt from SOEs toward the better-performing NSEs contributed to the larger part of the observed fall in average NPD ratios for the sample enterprises.
How were ªnancial resources allocated among the Chinese industrial enterprises during 1995–2002, which can be characterized as a period of high growth and steady reform? Which industries and regions were getting more ªnancial resources for their elite industrial enterprises? Tables 2 and 3 provide the answer. The two tables give us detailed information about credit allocation among China’s large and medium-sized industrial enterprises and illustrate the changing landscape of Chinese enter-prise ªnancing. In tables 2 and 3 the total debt for each industry or region are sorted
1 LINE SHORT REGULAR 1 LINE LONG
1 LINE SHORT REGULAR 1 LINE LONG T able 1. Distribution of total liabilities by ownership, 1995–2002 (RMB billion) 1995 1996 1997 1998 1999 2000 2001 2002 Change in total liabilities during 1995–2002 Change in total liabilities during 1995–2002 as percent-age of total in 1996 Amount of debt for Private 0 1 2 11 18 31 61 95 95 9,500.0% Collective 227 268 285 287 284 254 221 219 ⫺ 8 ⫺ 3.0% Mixed 231 279 380 514 690 1,033 1,424 1,642 1,41 1 505.7% For eign 152 232 285 329 371 400 561 619 467 201.3% Hong Kong, Macau, and T aiwan 164 190 214 276 297 305 359 389 225 118.4% State-owned 2,512 2,737 3,035 3,193 3,145 2,940 2,703 2,758 246 9.0% T otal 3,286 3,707 4,201 4,610 4,805 4,963 5,329 5,722 2,436 65.7% Shar e o f debt for Private 0.0% 0.0% 0.0% 0.2% 0.4% 0.6% 1.1% 1.7% ⫺ 1.7% Collective 6.9% 7.2% 6.8% 6.2% 5.9% 5.1% 4.1% 3.8% ⫺ 3.1% Mixed 7.0% 7.5% 9.0% 11.1% 14.4% 20.8% 26.7% 28.7% ⫺ 21.7% For eign 4.6% 6.3% 6.8% 7.1% 7.7% 8.1% 10.5% 10.8% ⫺ 6.2% Hong Kong, Macau, and T aiwan 5.0% 5.1% 5.1% 6.0% 6.2% 6.1% 6.7% 6.8% ⫺ 1.8% State-owned 76.4% 73.8% 72.2% 69.3% 65.5% 59.2% 50.7% 48.2% ⫺ 28.2% T otal 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%
by their amount in 2002, to make it easy to look for the winners and losers. The last two columns show the amount of change and the growth rate for total debts during the 1995–2002 period.
As shown in table 2, the top ªve industries in 2002, ranked by the level of their total debts, were electric power, steam and hot water; transport equipment manufactur-ing; smelting and pressing of ferrous metals; electronic and telecommunications equipment; and raw chemical materials and chemicals. The top ªve industries to-gether attracted RMB2.692 trillion in debt, or 47 percent of the total debt for the whole sample. The net gains in debt for the top ªve industries during 1995–2002 amounted to RMB1.465 trillion, or 60 percent of the gains by the whole sample. China’s ªnancial risks are heavily inºuenced by the performance of the above ªve sectors.
From the last column of table 2, the top ªve industries ranked by the growth of their total debts during 1995–2002 were tap water production and supply; electric power, steam, and hot water; electronic and telecommunications equipment; papermaking and paper products; and gas production and supply.
Clearly, the above leading industries, which have attracted investment in the last decade, are largely related to industrial infrastructure, intermediate inputs, raw ma-terials, production equipment, and utilities. Rapid development of these industries would lay a solid foundation for China’s further industrialization. In this sense, China’s enterprise ªnance looks increasingly driven by market forces. Of course, a risk-based regulation strategy would require extra attention to be paid to the sectors with heavy concentrations of investment. As we will see in the next section, some of the above sectors with rapid growth in enterprise debts do have high NPD ratios, especially the SOE-dominated utilities sector.
From table 3 we see that the top ªve regions ranked by the level of their total enter-prise debts in 2002 are Guangdong, Jiangsu, Shandong, Shanghai, and Liaoning. These regions are clearly becoming China’s new industrial centers. In section 5, we will examine region-speciªc enterprise performance, which is relevant for assessing debt risks across regions. Xiao (2005) examines enterprise performance in the north-east region of China in detail, and Xiao and Tu (2005) study China’s industrial pro-ductivity growth using the same set of data.
In the next section, we will show how much of the total debt shown in tables 1–3 is located in loss-making enterprises. Proªtability of enterprises becomes the crucial variable for our study. Reported proªts, however, present several problems. First, it
1 LINE SHORT REGULAR 1 LINE LONG
1 LINE SHORT REGULAR 1 LINE LONG T able 2. Distribution of total liabilities by industry , 1995–2002, by total liability in 2002 (RMB billion) 1995 1996 1997 1998 1999 2000 2001 2002 Change in total liabilities during 1995–2002 Change in total lia-bilities during 1995– 2002 as percentage o f total in 1995 [44] Electric power 296 317 361 521 614 719 802 934 638 215.5% [37] T ransport equipment 229 276 324 360 385 400 440 491 262 114.4% [32] Pr essing ferr ous 331 356 396 416 440 407 417 443 11 2 33.8% [41] Electr onic and telecommunications 145 167 198 229 251 286 366 424 279 192.4% [26] Raw chemicals 226 264 326 365 369 388 380 400 174 77.0% [17] T extile 258 274 287 275 244 234 230 234 ⫺ 24 ⫺ 9.3% [40] Electric equipment 136 164 185 195 194 195 218 226 90 66.2% [35] Or dinary machinery 157 180 194 207 202 205 209 222 65 41.4% [06] Coal mining 129 147 168 182 206 209 217 215 86 66.7% [31] Nonmetal pr oducts 138 166 186 195 201 195 209 215 77 55.8% [36] Special equipment 148 161 176 187 186 191 184 192 44 29.7% [25] Petr oleum pr ocessing 101 11 3 163 180 180 184 178 173 72 71.3% [33] Pr essing nonferr ous 95 101 11 6 132 140 140 153 162 67 70.5% [07] Petr oleum extract 151 157 164 170 160 158 149 153 2 1.3% [27] Medical 59 73 81 91 97 105 121 133 74 125.4% [22] Papermaking 53 65 72 73 80 88 121 125 72 135.8% [15] Beverage 73 83 99 103 108 11 1 11 1 114 41 56.2% [13] Food pr ocessing 87 107 11 2 114 106 101 103 11 1 24 27.6% [16] T obacco 70 76 82 72 71 74 109 108 38 54.3% [28] Chemical fiber 62 71 76 83 91 83 75 73 11 17.7% [34] Metal pr oducts 46 53 59 62 63 59 65 68 22 47.8% [14] Food pr oduction 37 41 42 46 48 48 56 65 28 75.7% [30] Plastic 31 38 43 46 48 49 55 61 30 96.8% [29] Rubber 36 43 48 50 51 49 52 51 15 41.7% [46] T ap water 14 15 23 27 30 34 39 46 32 228.6% [42] Instr uments 29 32 37 34 35 34 40 40 11 37.9% [18] Garments 18 23 24 26 29 33 38 39 21 116.7% [10] Nonmetal mining 16 16 18 19 23 28 23 29 13 81.3% [23] Printing 14 17 19 20 22 23 27 27 13 92.9% [19] Leather 16 19 21 21 21 21 23 24 8 50.0% [45] Gas pr oduction 10 13 13 16 18 20 19 23 13 130.0% [12] T imber logging 19 20 22 23 22 22 22 22 3 15.8% [20] T imber 10 12 16 16 18 19 21 22 12 120.0% [09] Nonferr ous mining 18 18 19 19 19 19 20 21 3 16.7% [43] Other manufacturing 9 11 11 10 12 11 13 13 4 44.4% [08] Ferr ous mining 6 7 8 11 9 8 8 10 4 66.7% [24] Cultural 5 8 8 9998 10 5 100.0% [21] Furnitur e 4 4 5 555662 50.0% T otal 3,282 3,708 4,202 4,610 4,807 4,964 5,327 5,725 2,443 74.4% Note: See appendix 2 for full industry names corr esponding to industry codes.
1 LINE SHORT REGULAR 1 LINE LONG T able 3. Distribution of total liabilities by region, 1995–2002, by total liability in 2002, (RMB billion) 1995 1996 1997 1998 1999 2000 2001 2002 Change o f total liabilities during 1995–2002 Change of total lia-bilities during 1995– 2002 as percentage o f total in 1995 [44] Guangdong 299 348 414 436 434 468 509 559 260 87.0% [32] Jiangsu 226 269 300 320 328 369 456 503 277 122.6% [37] Shandong 262 295 336 354 416 436 455 490 228 87.0% [31] Shanghai 231 273 323 337 347 334 349 376 145 62.8% [21] Liaoning 293 310 346 362 320 323 335 354 61 20.8% [50] Sichuan ⫹ Chongqing 183 199 233 240 319 295 303 325 142 77.6% [13] Hebei 143 166 184 199 213 224 249 257 11 4 79.7% [41] Henan 133 156 165 204 210 219 234 254 121 91.0% [42] Hubei 135 164 182 217 219 233 229 253 11 8 87.4% [23] Heilongjiang 149 149 167 190 183 191 189 195 46 30.9% [33] Zhejiang 121 143 153 157 159 166 170 187 66 54.5% [12] T ianjin 108 11 0 119 156 155 144 153 175 67 62.0% [22] Jilin 11 5 136 153 155 159 170 163 173 58 50.4% [1 1] Beijing 98 11 3 132 166 154 157 187 169 71 72.4% [61] Shaanxi 78 89 99 101 134 133 142 156 78 100.0% [35] Fujian 47 55 54 55 62 66 135 149 102 217.0% [43] Hunan 78 86 99 11 2 119 124 140 149 71 91.0% [34] Anhui 81 92 108 108 11 3 131 129 145 64 79.0% [14] Shanxi 78 89 100 129 129 135 126 133 55 70.5% [53] Y unnan 58 64 68 75 76 75 103 11 4 56 96.6% [62] Gansu 54 59 79 93 83 84 88 94 40 74.1% [45] Guangxi 60 62 71 80 85 90 89 90 30 50.0% [36] Jiangxi 58 68 72 78 87 88 90 88 30 51.7% [15] Inner Mongolia 51 55 63 71 78 72 83 87 36 70.6% [54] T ibet ⫹ Qinghai ⫹ Ningxia 34 34 43 56 57 63 55 79 45 132.4% [52] Guizhou 44 40 47 59 63 74 72 78 34 77.3% [65] Xinjiang 58 68 75 79 79 80 80 78 20 34.5% [46] Hainan 12 15 19 20 23 20 15 16 4 33.3% T otal 3,287 3,707 4,204 4,609 4,804 4,964 5,328 5,726 2,439 74.2%
is difªcult to check the consistency of reported proªts with enterprises’ other ªnancial variables because of China’s complicated accounting regulations. In other words, we do not know how reported proªts are calculated from other ªnancial variables reported in the NBS survey. Second, it is widely reported that enterprises sometimes manage their proªt numbers for many purposes, including legal or ille-gal tax evasion. For this paper, it seems useful to develop an alternative measure of proªtability, one that is based on a consistent set of ªnancial variables available from the NBS survey. Because the main purpose of the NBS survey is to calculate the value-added of industrial enterprises, it is possible to develop a measure of an en-terprise’s proªtability or potential proªtability based on the reconstructed compo-nents of its value-added.
We use the following variables available from the NBS survey to deªne the imputed proªtability of the sample enterprises:
VA⫽ value-added, including value-added taxes and ªnancial changes W⫽ wages and other employee compensation expenses
FC⫽ ªnancial charges, mainly interest payments D⫽ current depreciations
T⫽ all tax payments, including those for value-added taxes TA⫽ total assets
We can classify enterprises into eight proªtability groups: [⫺4]: if VA ≤ 0
[⫺3]: if VA ⫺ W ≤ 0 and VA ⬎ 0
[⫺2]: if VA ⫺ W ⫺ FC ≤ 0 and VA ⫺ W ⬎ 0
[⫺1]: if VA ⫺ W ⫺ FC ⫺ D ≤ 0 and VA ⫺ W ⫺ FC ⬎ 0
[⫹1]: if VA ⫺ W ⫺ FC ⫺ D ⫺ T ≤ 0 and VA ⫺ W ⫺ FC ⫺ D ⬎ 0
[⫹2]: if VA ⫺ W ⫺ FC ⫺ D ⫺ T ⬎ 0 and (VA ⫺ W ⫺ FC ⫺ D ⫺ T)/TA ≤ .05 [⫹3]: if (VA ⫺ W ⫺ FC ⫺ D ⫺ T)/TA ⬎ .05 and (VA ⫺ W ⫺ FC ⫺ D ⫺ T)/TA ≤ .15 [⫹4]: if (VA ⫺ W ⫺ FC ⫺ D ⫺ T)/TA ⬎ .15
Table 4 shows the number of enterprises in each of the eight proªtability groups over the period from 1995 to 2002. This imputed proªtability by group allows us to separate the NPDs into more disaggregated groups according to the qualitative and quantitative extent of loss making. Chinese banks are in the process of changing from four loan classiªcation categories (normal, overdue, doubtful, and poor) to the international standard of ªve categories (normal, special mention, substandard, doubtful, and loss). Unlike the classiªcation of bank loans, the proªtability 1 LINE SHORT
REGULAR 1 LINE LONG
1 LINE SHORT REGULAR 1 LINE LONG T able 4. Number of enterprises and share of total enterprises, by profitability , 1995–2002 Profitability 1995 1996 1997 1998 1999 2000 2001 2002 A v erage 1995–2002 [⫺ 4]: if VA ⱕ 0 867 1,006 969 901 569 458 450 602 728 [⫺ 3]: if VA ⫺ W ⱕ 0 and VA ⬎ 0 2,995 2,977 3,102 3,085 2,503 2,260 1,963 1,928 2,602 [⫺ 2]: if VA ⫺ W ⫺ FC ⱕ 0 and VA ⫺ W ⬎ 0 2,152 2,199 1,988 1,749 1,307 955 81 1 698 1,482 [⫺ 1]: if VA ⫺ W ⫺ FC ⫺ D ⱕ 0 and VA ⫺ W ⫺ FC ⬎ 0 1,815 1,757 1,839 1,943 1,841 1,684 1,740 1,724 1,793 [⫹ 1]: if VA ⫺ W ⫺ FC ⫺ D ⫺ T ⱕ 0 and VA ⫺ W ⫺ FC ⫺ D ⬎ 0 3,059 2,644 2,705 2,579 2,399 2,396 2,396 2,138 2,540 [⫹ 2]: if VA ⫺ W ⫺ FC ⫺ D ⫺ T ⬎ 0 and (VA ⫺ W ⫺ FC ⫺ D ⫺ T )/ TA ⱕ 0.05 5,095 5,106 5,336 5,284 5,292 5,129 5,225 4,985 5,182 [⫹ 3]: if (VA ⫺ W ⫺ FC ⫺ D ⫺ T )/ TA ⬎ 0.05 and (VA ⫺ W ⫺ FC ⫺ D ⫺ T )/ TA ⱕ 0.15 4,371 4,327 4,297 4,253 4,490 4,579 5,208 5,365 4,61 1 [⫹ 4]: if (VA ⫺ W ⫺ FC ⫺ D ⫺ T )/ TA ⬎ 0.15 2,189 2,958 2,721 2,499 3,062 3,277 4,105 4,780 3,199 T otal 22,543 22,974 22,957 22,293 21,463 20,738 21,898 22,220 22,136 [⫺ 4]: if VA ⱕ 0 3.8% 4.4% 4.2% 4.0% 2.7% 2.2% 2.1% 2.7% 3.3% [⫺ 3]: if VA ⫺ W ⱕ 0 and VA ⬎ 0 13.3% 13.0% 13.5% 13.8% 11.7% 10.9% 9.0% 8.7% 11.8% [⫺ 2]: if VA ⫺ W ⫺ FC ⱕ 0 and VA ⫺ W ⬎ 0 9.5% 9.6% 8.7% 7.8% 6.1% 4.6% 3.7% 3.1% 6.7% [⫺ 1]: if VA ⫺ W ⫺ FC ⫺ D ⱕ 0 and VA ⫺ W ⫺ FC ⬎ 0 8.1% 7.6% 8.0% 8.7% 8.6% 8.1% 7.9% 7.8% 8.1% [⫹ 1]: if VA ⫺ W ⫺ FC ⫺ D ⫺ T ⱕ 0 and VA ⫺ W ⫺ FC ⫺ D ⬎ 0 13.6% 11.5% 11.8% 11.6% 11.2% 11.6% 10.9% 9.6% 11.5% [⫹ 2]: if VA ⫺ W ⫺ FC ⫺ D ⫺ T ⬎ 0 and (VA ⫺ W ⫺ FC ⫺ D ⫺ T )/ TA ⱕ 0.05 22.6% 22.2% 23.2% 23.7% 24.7% 24.7% 23.9% 22.4% 23.4% [⫹ 3]: if (VA ⫺ W ⫺ FC ⫺ D ⫺ T )/T A ⬎ 0.05 and (VA ⫺ W ⫺ FC ⫺ D ⫺ T )/ TA ⱕ 0.15 19.4% 18.8% 18.7% 19.1% 20.9% 22.1% 23.8% 24.1% 20.8% [⫹ 4]: if (VA ⫺ W ⫺ FC ⫺ D ⫺ T )/ TA ⬎ 0.15 9.7% 12.9% 11.9% 11.2% 14.3% 15.8% 18.7% 21.5% 14.5% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% Note: VA value-added, including value-added taxes and financial changes; W wages and other employee compensation expenses; FC financial char ges, mainly inter est payments; D curr ent depr ecia-tions; T all tax payments including value-added taxes; T A total assets.
classiªcation proposed here reveals the underlying economic conditions, for example:
• Enterprises in proªtability group [⫺4] create negative value-added. They should be closed immediately according to economic principles. The quality of their debts is the worst among the eight groups by proªtability.
• Enterprises in group [⫺3] have positive value-added but cannot pay all of their wage bills. In economics, they cannot even cover their variable costs. They should also be closed as soon as possible to avoid incurring new losses. The quality of their debts will worsen every day as the losses accumulate.
• Enterprises in group [⫺2] can pay their wage bills but cannot pay all of their ªnancial charges. The quality of their debts is poor, but because their investment is sunk, such ªrms may have reasons to continue operations in the short run to maintain employment while waiting for a turnaround after reorganization. • Enterprises in group [⫺1] can pay their wage bills and ªnancial charges but
can-not cover all of their depreciation charges. The quality of their debts will fall as capital is depleted.
We will leave the more detailed analysis of NPDs based on the above proªtability classiªcations for a separate paper. Here we focus on the big picture ªrst and clas-sify enterprises in the ªrst four groups as loss making and the last four groups as proªt making, based on imputed proªtability.
Table 5 shows the number of enterprises making proªts or losses based on both re-ported and imputed proªts over the period from 1995 to 2002. The number of loss-making enterprises by imputed proªtability was stable at about 8,000 (34–35 per-cent) during 1995–98 and fell rapidly afterward to 4,952 (22.3 perper-cent) in 2002. The number of loss-making enterprises by reported proªtability was 6,937 (30.8 percent) in 1995 and rose sharply to 8,987 (40.3 percent) in 1998, then dropped to 6,295 (or 28.3 percent) in 2002. In the next section, we will use both imputed and reported proªtability to estimate the amount and ratio of NPD. Although the two proªt-ability measurements are different in concept and measurement, both are useful for assessing the quality of enterprise debt. Imputed proªtability is more useful for comparing enterprise performance across groups because it is based on a consistent set of reported ªnancial variables, but it is different from actually reported
proªtability. Imputed proªts can be larger than reported proªts for several reasons. First, when output is not sold or is still in inventory, some of the value-added might not turn into actual proªts. Second, reported proªts are likely to be lower than im-puted proªts as a result of legal or illegal tax evasion or proªt hiding. In other re-1 LINE SHORT
REGULAR 1 LINE LONG
lated papers (Liu and Xiao 2004; Cai, Liu, and Xiao 2005), we examine the issue of proªt disguising in detail.
3. Estimated level and ratio of NPD
Using the method developed in section 2, we report the main NPD statistics for the whole sample as well as for groups classiªed by ownership, industry, and region. Table 6 shows the amount of NPD as well as the NPD ratio for the whole sample during the 1995–2002 period. There are two sets of NPD statistics in the table: the upper part is derived from imputed proªts and the lower part from reported proªts. The amount and ratio of NPD are calculated separately for three categories of debts: total liabilities, long-term liabilities, and short-term liabilities. They are similar in size and trend, with the NPD ratio for short-term liabilities declining slightly faster than the ratio for long-term liabilities.
According to imputed proªtability, the NPD ratio for the whole sample was stable at around 27–30 percent during 1995–99 but declined rapidly afterward to only 18.4 percent in 2002, with the amount of NPD at about RMB1 trillion.
According to reported proªtability, the NPD ratio for the whole sample was 24.1 percent in 1995, rising to 34.3 percent in 1998 and then falling to 22.9 percent in 2002, with the amount of NPDs at about RMB1.3 trillion. According to the CBRC, China’s
NPL ratio fell sharply to 19.6 percent, with the amount of NPLs at RMB2.5 trillion 1 LINE SHORT REGULAR 1 LINE LONG Table 5. Number and share of enterprises making profits or losses, 1995–2002
1995 1996 1997 1998 1999 2000 2001 2002
Average 1995–2002
Based on imputed profits
Number of enterprises Making profits 14,714 15,035 15,059 14,615 15,243 15,381 16,934 17,268 15,531 Making losses 7,829 7,939 7,898 7,678 6,220 5,357 4,964 4,952 6,605 Total 22,543 22,974 22,957 22,293 21,463 20,738 21,898 22,220 22,136 Share Making profits 65.3% 65.4% 65.6% 65.6% 71.0% 74.2% 77.3% 77.7% 70.2% Making losses 34.7% 34.6% 34.4% 34.4% 29.0% 25.8% 22.7% 22.3% 29.8% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%
Based on reported profits
Number of enterprises Making profits 15,606 15,088 14,476 13,306 14,328 15,144 15,510 15,925 14,923 Making losses 6,937 7,886 8,481 8,987 7,135 5,594 6,388 6,295 7,213 Total 22,543 22,974 22,957 22,293 21,463 20,738 21,898 22,220 22,136 Share Making profits 69.2% 65.7% 63.1% 59.7% 66.8% 73.0% 70.8% 71.7% 67.4% Making losses 30.8% 34.3% 36.9% 40.3% 33.2% 27.0% 29.2% 28.3% 32.6% Total 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%
1 LINE SHORT REGULAR 1 LINE LONG T able 6. Amount of nonperforming debt and nonperforming debt ratio for whole sample, 1995–2002 1995 1996 1997 1998 1999 2000 2001 2002 Change in 1995–2002 Change in 1998–2002 Based on imputed profits Amount of nonperforming debt (RMB billion) W ithin total liabilities 914 1,095 1,204 1,371 1,315 1,128 1,091 1,051 137 ⫺ 320 W ithin long-term liabilities 308 370 389 470 472 395 365 343 35 ⫺ 127 W ithin short-term liabilities 605 724 812 895 836 727 716 709 104 ⫺ 186 Nonperforming debt ratio (per cent) W ithin total liabilities 27.8 29.5 28.7 29.7 27.4 22.7 20.5 18.4 ⫺ 9.5 ⫺ 11.4 W ithin long-term liabilities 26.6 29.5 28.2 30.3 29.2 24.2 21.9 19.9 ⫺ 6.7 ⫺ 10.4 W ithin short-term liabilities 28.5 29.6 29.1 29.5 26.5 22.2 19.8 17.9 ⫺ 10.6 ⫺ 11.6 Based on reported profits Amount of nonperforming debt (RMB billion) W ithin total liabilities 792 1,002 1,167 1,580 1,427 1,185 1,392 1,31 1 519 ⫺ 269 W ithin long-term liabilities 282 327 368 556 504 41 1 441 408 126 ⫺ 148 W ithin short-term liabilities 509 674 793 1,014 916 759 930 891 382 ⫺ 123 Nonperforming debt ratio (per cent) W ithin total liabilities 24.1 27.0 27.8 34.3 29.7 23.9 26.1 22.9 ⫺ 1.2 ⫺ 11.4 W ithin long-term liabilities 24.4 26.1 26.7 35.9 31.1 25.1 26.5 23.7 ⫺ 0.7 ⫺ 12.3 W ithin short-term liabilities 24.0 27.5 28.4 33.4 29.0 23.2 25.8 22.5 ⫺ 1.5 ⫺ 11.0
by the middle of 2003. Given the different deªnitions of NPL and NPD, the results for the NPD statistics look consistent with the CBRC statistics for NPLs. In the next section, we will further examine the trend of NPD ratios for the whole sample. We compare the NPD statistics for different types of enterprises by ownership, where NPD is derived from both imputed proªtability (table 7) and reported proªtability (table 8). These tables show that NPD ratios vary signiªcantly across types of enterprises by ownership and that SOEs have much higher NPD ratios than NSEs. In 2002, the NPD ratio for SOEs was 25.4 percent by imputed proªtability and 25.8 percent by reported proªtability. The NPD ratio for private enterprises was 7.4 percent by imputed proªtability and 15.8 percent by reported proªtability. The NPD ratio for domestic mixed-ownership enterprises was 10.8 percent by imputed proªtability and 20.2 percent by reported proªtability.
From the NPD statistics in tables 7 and 8, it is possible to decompose the fall of the average NPD ratio for the whole sample into two parts: one resulting from improve-ment in NPD ratios in each type of enterprise and the other resulting from the redis-tribution of debts from SOEs to the better-performing NSEs.
1 LINE SHORT REGULAR 1 LINE LONG Table 7. Amount of nonperforming debt and nonperforming debt ratio, estimated from
imputed profitability, by ownership, 1995–2002
1995 1996 1997 1998 1999 2000 2001 2002
Amount of nonperforming debt estimated from imputed profitability (RMB billion)
Private 0 0 0 2 2 4 5 7
Collective 53 55 60 50 47 32 25 22
Mixed 39 63 90 107 165 191 207 178
Foreign 36 69 77 94 74 63 98 94
Hong Kong, Macau, and Taiwan 26 32 37 68 54 53 53 49
State-owned 761 876 938 1,049 972 786 702 701
Total 915 1,095 1,202 1,370 1,314 1,129 1,090 1,051
Nonperforming ratio estimated from imputed profitability (percent)
Private 0.0 0.0 0.0 18.2 11.8 12.9 8.2 7.4
Collective 23.3 20.6 21.1 17.4 16.5 12.6 11.3 10.0
Mixed 16.9 22.5 23.7 20.8 23.9 18.5 14.5 10.8
Foreign 23.7 29.7 27.0 28.7 20.0 15.8 17.5 15.2
Hong Kong, Macau, and Taiwan 15.9 16.8 17.3 24.6 18.2 17.4 14.8 12.6
State-owned 30.3 32.0 30.9 32.9 30.9 26.7 26.0 25.4
Total 27.8 29.5 28.6 29.7 27.4 22.7 20.5 18.4
Share of nonperforming debt estimated from imputed profitability (percent)
Private 0.0 0.0 0.0 0.1 0.2 0.4 0.5 0.7
Collective 5.8 5.0 5.0 3.6 3.6 2.8 2.3 2.1
Mixed 4.3 5.8 7.5 7.8 12.6 16.9 19.0 16.9
Foreign 3.9 6.3 6.4 6.9 5.6 5.6 9.0 8.9
Hong Kong, Macau, and Taiwan 2.8 2.9 3.1 5.0 4.1 4.7 4.9 4.7
State-owned 83.2 80.0 78.0 76.6 74.0 69.6 64.4 66.7
Let us assume that Ri t
is the NPD ratio in year t for enterprise group i and Si t
is the share of debts for group i in year t. Then the NPD ratio for the whole sample in year tcan be calculated from the following formula:
NPD ratio⫽ Σi0 5 Ri Ri Si Si 2002 1995 2002 1995 . ∗( + ) (∗ − )+ NPD ratio⫽Σi0 5 Σi Ri Ri Si Si 2002 1995 2002 1995 . ∗ ( − ) (∗ + ),
where i⫽ private enterprises, collective enterprises, mixed enterprises, foreign en-terprises, Hong Kong, Macau, or Taiwanese enterprises (HK-M-Taiwan), or SOEs. The change in NPD for the whole sample from 1995 to 2002 can be presented equiv-alently in the following formats:
R2002⫺ R1995⫽ Σ Σ iRi Si iRi Si 2002∗ 2002− 1995∗ 1995 R2002⫺ R1995⫽Σ i0 5 Ri Ri Si Si 2002 1995 2002 1995 . ∗( + ) (∗ − )+ R2002⫺ R1995⫽Σ i0 5 Ri Ri Si Si 2002 1995 1995 2002 . ∗( − ) (∗ + ).
The ªrst term in the above equation is the ªrst component of the change in the NPD ratio for the whole sample during 1995–2002 that can be attributed to the shift of the 1 LINE SHORT
REGULAR 1 LINE LONG
Table 8. Amount of nonperforming debt and nonperforming debt ratio, estimated from reported profitability, by ownership 1995–2002
1995 1996 1997 1998 1999 2000 2001 2002
Amount of nonperforming debt estimated from reported profitability (RMB billion)
Private 0 0 0 4 4 7 12 15
Collective 53 67 74 84 69 52 48 42
Mixed 28 37 69 136 172 214 296 332
Foreign 44 82 102 136 107 98 177 139
Hong Kong, Macau, and Taiwan 29 45 52 89 74 63 77 71
State-owned 638 770 869 1,131 999 752 782 712
Total 792 1,001 1,166 1,580 1,425 1,186 1,392 1,311
Nonperforming debt ratio estimated from reported profitability (percent)
Private 0.0 0.0 0.0 36.4 23.5 22.6 19.4 15.8
Collective 23.3 25.0 26.1 29.3 24.3 20.5 21.7 19.2
Mixed 12.2 13.3 18.2 26.5 24.9 20.7 20.8 20.2
Foreign 28.9 35.3 35.8 41.3 28.9 24.5 31.6 22.4
Hong Kong, Macau, and Taiwan 17.7 23.7 24.3 32.4 24.9 20.7 21.4 18.2
State-owned 25.4 28.1 28.6 35.4 31.8 25.6 28.9 25.8
Total 24.1 27.0 27.8 34.3 29.7 23.9 26.1 22.9
Share of nonperforming debt estimated from reported profitability (percent)
Private 0.0 0.0 0.0 0.3 0.3 0.6 0.9 1.1
Collective 6.7 6.7 6.3 5.3 4.8 4.4 3.4 3.2
Mixed 3.5 3.7 5.9 8.6 12.1 18.0 21.3 25.3
Foreign 5.6 8.2 8.7 8.6 7.5 8.3 12.7 10.6
Hong Kong, Macau, and Taiwan 3.7 4.5 4.5 5.6 5.2 5.3 5.5 5.4
State-owned 80.6 76.9 74.5 71.6 70.1 63.4 56.2 54.3
total liabilities across ownership groups while holding the individual ownership group’s NPD ratio at its average level for 1995 and 2002. Using statistics from tables 1, 7, and 8, this ªrst component is⫺3.86 percent for the imputed-proªtability method and⫺2.33 percent for the reported-proªtability method.
The second term in the NPD equation is the component of change in the NPD ratio for the whole sample during 1995–2002 that can be attributed to the fall in individ-ual ownership groups’ NPD ratio while holding constant the distribution of total li-abilities across ownership groups at its average level for 1995 and 2002. This second component is⫺4.74 percent for the imputed-proªtability method and 1.13 percent for the reported-proªtability method.
Hence, according to the imputed-proªtability method, the NPD ratio for the whole sample fell from 27.8 percent in 1995 to 18.4 percent in 2002, a drop of 9.4 percentage points. Out of these 9.4 percentage points, 3.86 percentage points can be attributed to the shift of ªnancial resources from SOEs to the better-performing NSEs, which have lower NPD ratios than SOEs.
According to the reported-proªtability method, the NPD ratio for the whole sample fell only slightly, from 24.1 percent in 1995 to 22.9 percent in 2002, a drop of only 1.2 percentage points. The decomposing of this 1.2 percentage points shows that the shift of ªnancial resources from SOEs to the better-performing NSEs led to a 2.33-percentage-point drop in the NPD ratio for the whole sample, whereas the changes in the NPD ratios for individual ownership groups led to an increase of 1.13 per-centage points in the NPD ratio for the whole sample.
Clearly, the decline in the NPD ratio is more signiªcant according to the imputed-proªtability method, compared to the reported-imputed-proªtability method. As previously noted, we are not clear how reported proªts are calculated because of large varia-tions in accounting and proªt-reporting practices across types of enterprises, but we know exactly how imputed proªts are calculated from the ªnancial variables that are used for measuring GDP. We think both measures are useful. The NPD statistics derived from imputed proªtability can be used for comparing the underlying per-formance of different groups of enterprises, and the NPD statistics from reported proªtability better reºect the actual outcomes that creditors are going to face when they deal with enterprises.
Tables 9–12 present NPD statistics by industry for 1995–2002. The results shown in tables 9 and 10 are derived using imputed proªtability, and the results in tables 11 and 12 are derived from reported proªtability. Tables 13–16 present NPD statistics
1 LINE SHORT REGULAR 1 LINE LONG
by region during 1995–2002. Tables 13 and 14 are derived from imputed proªtability, and tables 15 and 16 are derived from reported proªtability. Tables 9–16 are sorted by the results in the last column (for 2002) so that readers can easily see the best and worst performers in the quality of enterprise debts by region.
The information here shows the big picture on the quality of enterprise debts across industry and region and can be used by policymakers, as well as by banks, inves-tors, and enterprises as a benchmark against which to check the performance of their own debt portfolios. This information is a public good and contributes to more scientiªc management of debt risks in China. Bankers from Shanghai and
Guangdong might want to know the NPD statistics in their regions. Ofªcials in charge of utilities might want to know how bad that sector’s enterprise debts are 1 LINE SHORT
REGULAR 1 LINE LONG
Table 9. Amount of nonperforming debt by industry, estimated from imputed profitability, 1995–2002, by nonperforming debt in 2002 (RMB billion)
1995 1996 1997 1998 1999 2000 2001 2002 [44] Electric power 52 86 80 143 188 169 174 177 [26] Raw chemicals 52 65 92 127 108 88 97 89 [06] Coal mining 63 76 77 81 112 102 67 77 [36] Special equipment 69 74 86 103 98 93 83 76 [37] Transport equipment 63 81 90 101 96 93 103 72
[41] Electronic and telecommunications 47 52 60 64 54 56 52 60
[31] Nonmetal products 42 56 72 65 59 49 48 51 [35] Ordinary machinery 52 71 74 81 74 65 62 51 [17] Textile 113 124 106 105 72 47 44 44 [40] Electric equipment 39 51 58 53 52 45 48 40 [22] Papermaking 13 13 22 28 21 19 25 27 [28] Chemical fiber 17 22 21 33 26 22 24 24 [32] Pressing ferrous 75 70 85 81 85 62 35 23 [27] Medical 13 20 19 20 15 11 13 19 [46] Tap water 7 7 12 9 10 12 13 19 [13] Food processing 32 42 41 42 32 21 17 18 [33] Pressing nonferrous 28 35 36 47 28 22 22 16 [42] Instruments 15 17 19 17 14 13 14 16 [14] Food production 13 13 16 14 11 8 12 14 [34] Metal products 16 19 23 27 22 18 13 14 [45] Gas production 9 12 11 10 10 13 13 14 [15] Beverage 11 12 13 14 12 13 9 13 [07] Petroleum extract 3 4 13 0 13 11 18 12 [25] Petroleum processing 3 9 7 15 18 8 17 10 [10] Nonmetal mining 7 5 7 7 6 6 7 9 [29] Rubber 9 9 9 11 18 9 9 9 [30] Plastic 12 12 14 16 14 10 7 9 [09] Nonferrous mining 5 6 6 9 6 5 6 7 [12] Timber logging 4 4 6 7 9 8 7 7 [20] Timber 5 4 6 8 7 6 5 6 [18] Garments 4 5 4 7 5 4 6 5 [23] Printing 4 4 6 5 6 6 6 5 [08] Ferrous mining 5 4 1 5 2 2 1 3 [19] Leather 6 5 5 7 4 5 3 3 [43] Other manufacturing 3 2 3 3 3 3 3 3 [24] Cultural 1 2 1 3 2 3 2 2 [16] Tobacco 1 1 1 3 1 2 3 1 [21] Furniture 2 1 1 1 2 2 2 1 Total 915 1,095 1,203 1,372 1,315 1,131 1,090 1,046
compared with those of other industries. These patterns of NPD ratios at the aggre-gate levels by ownership, industry and region are useful for illustrating the overall quality and distribution of enterprise debts in China as well as for contributing to informed policy debates.
4. Patterns of NPDs
By applying a simple regression method to the disaggregated NPD ratios, we can summarize the variability in NPD ratios for two relevant dimensions: one is the de-clining trend in NPD ratios and the other is the gap in NPD ratios across ownership type, industry, and region. Tables 17–19 show the results of six regressions using the group NPD ratios reported in tables 7 and 8, 9 and 11, and 13 and 15, respectively. In
1 LINE SHORT REGULAR 1 LINE LONG Table 10. Nonperforming debt ratios by industry, estimated from imputed profitability,
1995–2002, by nonperforming debt ratio in 2002 (percent)
1995 1996 1997 1998 1999 2000 2001 2002 [45] Gas production 90.0 92.3 84.6 58.8 58.8 68.4 72.2 60.9 [46] Tap water 50.0 43.8 50.0 34.6 32.3 34.3 33.3 41.3 [36] Special equipment 46.3 46.0 48.9 55.1 52.7 48.9 45.1 39.6 [42] Instruments 51.7 53.1 51.4 50.0 40.0 38.2 35.0 39.0 [06] Coal mining 49.2 51.7 45.8 44.5 54.6 48.8 30.9 35.8 [09] Nonferrous mining 27.8 35.3 31.6 47.4 33.3 26.3 30.0 35.0 [28] Chemical fiber 27.4 31.0 28.0 39.8 28.9 26.5 32.0 32.9 [12] Timber logging 21.1 20.0 27.3 30.4 40.9 34.8 33.3 31.8 [10] Nonmetal mining 41.2 33.3 38.9 36.8 27.3 21.4 30.4 31.0 [08] Ferrous mining 83.3 57.1 12.5 45.5 25.0 22.2 12.5 30.0 [20] Timber 55.6 33.3 37.5 50.0 38.9 31.6 23.8 28.6 [31] Nonmetal products 30.4 33.7 38.7 33.3 29.5 25.1 23.0 23.8 [43] Other manufacturing 30.0 20.0 27.3 30.0 25.0 27.3 25.0 23.1 [35] Ordinary machinery 33.1 39.7 38.1 39.1 36.6 31.7 29.5 23.0 [26] Raw chemicals 23.0 24.6 28.3 34.8 29.3 22.7 25.5 22.3 [22] Papermaking 24.5 20.3 30.6 38.4 26.3 21.6 20.7 21.6 [14] Food production 35.1 31.7 38.1 29.8 22.9 16.7 21.1 21.5 [34] Metal products 34.8 35.8 39.0 43.5 34.9 30.5 20.0 20.6 [24] Cultural 20.0 25.0 12.5 33.3 22.2 33.3 25.0 20.0 [44] Electric power 17.6 27.1 22.2 27.4 30.6 23.5 21.7 19.0 [17] Textile 44.0 45.3 36.9 38.2 29.5 20.0 19.1 18.8 [23] Printing 28.6 25.0 31.6 25.0 26.1 26.1 22.2 18.5 [29] Rubber 25.0 21.4 18.8 22.0 35.3 18.4 17.3 18.0 [40] Electric equipment 28.5 31.1 31.4 27.2 26.8 23.1 22.0 17.7 [21] Furniture 50.0 25.0 20.0 20.0 40.0 40.0 33.3 16.7 [13] Food processing 36.4 39.3 36.6 36.5 30.2 20.8 16.5 16.2 [30] Plastic 38.7 31.6 32.6 34.8 29.2 20.4 12.7 15.0 [37] Transport equipment 27.5 29.3 27.7 28.1 24.9 23.3 23.4 14.7 [27] Medical 22.0 27.4 23.5 22.2 15.5 10.5 10.7 14.3
[41] Electronic and telecommunications 32.2 31.1 30.3 27.9 21.5 19.6 14.2 14.2
[18] Garments 23.5 21.7 16.7 25.9 16.7 12.5 16.2 13.2 [19] Leather 37.5 26.3 23.8 33.3 19.0 23.8 13.0 12.5 [15] Beverage 15.1 14.5 13.1 13.7 11.1 11.7 8.1 11.4 [33] Pressing nonferrous 29.5 34.7 30.8 35.6 20.0 15.7 14.4 9.9 [07] Petroleum extract 2.0 2.5 7.9 0.0 8.1 7.0 12.1 7.8 [25] Petroleum processing 3.0 7.9 4.3 8.3 10.0 4.3 9.6 5.8 [32] Pressing ferrous 22.7 19.7 21.5 19.5 19.3 15.2 8.4 5.2 [16] Tobacco 1.4 1.3 1.2 4.2 1.4 2.7 2.8 0.9 Total 27.9 29.6 28.6 29.8 27.4 22.8 20.5 18.3
each of the six regressions, the independent variables include a time trend (year) and a categorical variable (ownership, industry, or region). Each categorical variable has the “whole sample” dummy to match the NPD ratio for the whole sample. The regression equations can be written as
NPD ratio⫽ f (year, categorical variable).
We use weighted regressions to discount the impact of the NPD ratios in the early years. (The weights used are listed in the footnotes of tables 17, 18, and 19.) The re-gression coefªcients for the time trend variable (year) indicate how fast the NPD ra-tio would fall every year based on the variability of the NPD rara-tios reported for each group in the relevant tables. In principle, the declining trend of NPD ratios for all 1 LINE SHORT
REGULAR 1 LINE LONG
Table 11. Amount of nonperforming debt by industry, estimated from reported profitability, 1995–2002, by nonperforming debt in 2002 (RMB billion)
1995 1996 1997 1998 1999 2000 2001 2002
[44] Electric power 83 39 76 119 161 147 173 183
[26] Raw chemicals 39 51 103 174 158 123 134 140
[37] Transport equipment 57 68 89 106 114 117 108 117
[25] Petroleum processing 4 22 16 76 39 86 98 76
[41] Electronic and telecommunications 31 41 53 60 54 35 59 73
[31] Nonmetal products 43 64 83 88 74 54 70 72 [17] Textile 104 135 128 135 88 51 74 69 [35] Ordinary machinery 36 52 58 72 64 59 68 65 [36] Special equipment 50 57 68 88 81 76 86 65 [40] Electric equipment 30 41 52 61 52 42 62 45 [28] Chemical fiber 11 20 16 34 27 17 32 32 [33] Pressing nonferrous 18 35 35 68 46 21 33 32 [13] Food processing 29 52 53 68 51 28 29 31 [22] Papermaking 13 17 26 32 27 23 40 30 [15] Beverage 18 21 22 29 27 25 27 29 [32] Pressing ferrous 45 61 69 68 81 41 43 25 [06] Coal mining 29 33 31 79 99 78 67 23 [27] Medical 12 17 24 23 15 14 17 20 [34] Metal products 15 21 26 30 27 21 19 19 [14] Food production 12 15 17 18 13 12 17 18 [30] Plastic 11 12 14 17 14 13 13 15 [46] Tap water 4 5 10 12 9 9 11 15 [16] Tobacco 5 6 7 6 5 8 25 14 [42] Instruments 10 12 15 17 14 9 11 13 [10] Nonmetal mining 7 6 6 7 6 7 5 11 [29] Rubber 11 11 10 13 19 18 15 11 [45] Gas production 4 10 9 8 8 10 8 10 [07] Petroleum extract 28 40 6 22 7 2 3 9 [18] Garments 3 5 7 8 6 5 9 8 [20] Timber 4 5 7 8 7 6 8 8 [19] Leather 5 6 6 6 5 5 5 7 [09] Nonferrous mining 7 8 6 9 6 3 5 6 [12] Timber logging 3 5 8 7 7 8 7 6 [23] Printing 3 3 5 5 4 3 4 4 [08] Ferrous mining 5 3 2 2 3 3 2 3 [43] Other manufacturing 2 3 3 3 4 3 3 3 [21] Furniture 1 1 2 1 1 1 2 1 [24] Cultural 1 1 1 2 1 1 1 1 Total 793 1,004 1,169 1,581 1,424 1,184 1,393 1,309
the groups is related to the improvement of the general market environment of the Chinese economy as a result of reform and opening. The regression coefªcients for the categorical variable indicate the average gap between the NPD ratio of that par-ticular category and the NPD ratio of the base category (which is indicated by a zero value for the coefªcient and a blank value for the t-statistics in the tables) after re-moving the inºuence of the declining trend in NPD ratio. The negative sign means “lower than” the NPD ratio of the base category.
For example, table 17 shows that based on the NPD ratios estimated from imputed proªtability (reported in table 7), the NPD ratio for a particular group is likely to de-cline on average by 1.5 percentage points each year. For private enterprises, NPD ra-tios in a given year are likely to be 21.3 percentage points lower than those for SOEs
1 LINE SHORT REGULAR 1 LINE LONG Table 12. Nonperforming debt ratios by industry, estimated from reported profitability,
1995–2002, by nonperforming debt ratio in 2002 (percent)
1995 1996 1997 1998 1999 2000 2001 2002 [25] Petroleum processing 4.0 19.3 9.8 42.2 21.7 46.7 54.7 43.9 [28] Chemical fiber 17.7 28.2 21.1 41.0 29.7 20.5 42.1 43.8 [45] Gas production 40.0 76.9 69.2 50.0 44.4 50.0 44.4 41.7 [10] Nonmetal mining 43.8 37.5 33.3 35.0 27.3 25.9 21.7 37.9 [20] Timber 40.0 41.7 43.8 50.0 41.2 31.6 38.1 36.4 [26] Raw chemicals 17.3 19.3 31.6 47.7 42.8 31.7 35.3 34.9 [36] Special equipment 33.8 35.6 38.6 47.1 43.5 39.8 46.7 33.9 [31] Nonmetal products 31.2 38.3 44.9 45.4 37.0 27.7 33.5 33.6 [46] Tap water 26.7 31.3 43.5 44.4 29.0 26.5 28.2 32.6 [42] Instruments 34.5 37.5 40.5 51.5 40.0 26.5 27.5 31.7 [08] Ferrous mining 71.4 37.5 25.0 18.2 33.3 37.5 25.0 30.0 [17] Textile 40.3 49.3 44.6 49.1 35.9 21.8 32.0 29.5 [35] Ordinary machinery 22.8 28.9 29.9 34.8 31.7 28.8 32.5 29.4 [19] Leather 31.3 31.6 28.6 28.6 23.8 22.7 21.7 29.2 [09] Nonferrous mining 38.9 47.1 31.6 47.4 33.3 15.8 25.0 28.6 [34] Metal products 31.9 38.9 44.1 48.4 42.9 35.6 29.2 28.4 [13] Food processing 33.3 48.6 47.7 59.1 48.1 28.0 28.2 27.9 [14] Food production 32.4 36.6 40.5 39.1 27.1 24.5 30.4 27.7 [12] Timber logging 15.8 25.0 36.4 30.4 31.8 34.8 31.8 26.1 [15] Beverage 24.7 25.3 22.2 28.4 25.2 22.5 24.5 25.4 [30] Plastic 35.5 31.6 32.6 37.0 29.2 26.5 23.6 24.6 [22] Papermaking 24.5 26.2 36.1 43.8 33.8 26.4 33.1 24.0 [37] Transport equipment 24.9 24.6 27.4 29.4 29.6 29.2 24.5 23.8 [29] Rubber 30.6 25.6 20.8 25.5 37.3 36.7 28.8 21.6 [43] Other manufacturing 22.2 27.3 27.3 33.3 33.3 27.3 23.1 21.4 [18] Garments 17.6 21.7 29.2 29.6 20.7 15.2 23.7 20.5 [40] Electric equipment 22.1 25.0 28.1 31.3 26.8 21.5 28.4 19.9 [33] Pressing nonferrous 18.9 34.7 30.2 51.5 32.9 15.1 21.4 19.8 [44] Electric power 28.0 12.3 21.1 22.8 26.2 20.5 21.6 19.6
[41] Electronic and telecommunications 21.4 24.6 26.6 26.2 21.5 12.2 16.1 17.2
[21] Furniture 25.0 33.3 40.0 20.0 20.0 20.0 33.3 16.7 [27] Medical 20.3 23.6 29.3 25.3 15.5 13.3 14.0 15.0 [23] Printing 21.4 17.6 25.0 23.8 18.2 13.0 14.8 14.8 [16] Tobacco 7.0 7.9 8.5 8.3 7.0 10.8 22.9 13.0 [24] Cultural 16.7 12.5 12.5 22.2 11.1 11.1 12.5 11.1 [06] Coal mining 22.5 22.3 18.5 43.4 48.1 37.3 30.9 10.7 [07] Petroleum extract 18.5 25.5 3.7 12.9 4.4 1.3 2.0 5.9 [32] Pressing ferrous 13.6 17.1 17.5 16.3 18.4 10.1 10.3 5.6 Total 24.1 27.1 27.8 34.3 29.6 23.9 26.1 22.9
1 LINE SHORT REGULAR 1 LINE LONG
1995–2002, by nonperforming debt in 2002 (RMB billion)
1995 1996 1997 1998 1999 2000 2001 2002 [31] Shanghai 70 94 95 98 92 78 69 81 [44] Guangdong 64 90 113 131 106 90 86 80 [41] Henan 23 39 37 63 61 64 69 72 [23] Heilongjiang 60 58 56 69 71 62 56 58 [21] Liaoning 105 117 115 110 87 63 75 57 [50] Sichuan⫹ Chongqing 71 71 109 88 127 124 104 57 [12] Tianjin 52 32 38 61 53 24 39 56 [42] Hubei 27 53 39 36 34 36 35 46 [13] Hebei 33 45 41 42 47 45 49 45 [37] Shandong 47 48 50 64 69 58 44 45 [11] Beijing 27 45 67 99 68 64 55 42 [22] Jilin 36 35 38 39 35 32 39 41 [32] Jiangsu 39 49 57 57 59 51 49 40 [61] Shaanxi 37 49 49 46 48 38 46 38 [14] Shanxi 24 33 29 50 44 33 24 35
[54] Tibet⫹ Qinghai ⫹ Ningxia 12 18 23 18 14 11 11 34
[43] Hunan 29 28 33 37 35 34 30 32 [52] Guizhou 16 15 15 28 25 28 26 31 [34] Anhui 26 27 26 35 31 44 14 23 [45] Guangxi 18 18 23 20 21 17 16 22 [15] Inner Mongolia 17 16 14 27 46 25 34 21 [53] Yunnan 9 11 20 24 23 17 20 21 [35] Fujian 7 10 11 11 6 8 34 20 [62] Gansu 12 22 29 30 28 26 15 15 [36] Jiangxi 20 26 29 30 32 19 18 12 [33] Zhejiang 19 25 24 37 25 18 17 11 [65] Xinjiang 9 13 14 16 21 17 15 11 [46] Hainan 4 7 9 5 8 2 4 3 Total 913 1,094 1,203 1,371 1,316 1,128 1,093 1,049
Table 14. Nonperforming debt ratios by region, estimated from imputed profitability, 1995– 2002, by nonperforming debt ratio in 2002 (percent)
1995 1996 1997 1998 1999 2000 2001 2002
[54] Tibet⫹ Qinghai ⫹ Ningxia 35.3 52.9 53.5 31.6 24.6 17.7 20.0 42.5
[52] Guizhou 36.4 37.5 31.9 47.5 39.7 37.8 36.1 40.3 [12] Tianjin 48.1 29.4 32.2 39.1 34.2 16.6 25.5 32.0 [23] Heilongjiang 40.0 38.9 33.7 36.3 38.8 32.6 29.5 29.7 [41] Henan 17.3 25.0 22.3 30.9 29.0 29.2 29.5 28.3 [14] Shanxi 30.4 37.1 29.0 38.5 34.1 24.4 19.0 26.5 [11] Beijing 27.6 39.8 51.1 59.6 44.2 40.8 29.6 24.9 [45] Guangxi 30.5 29.5 32.9 24.7 24.7 18.9 18.0 24.4 [61] Shaanxi 47.4 55.7 50.0 45.1 35.8 28.6 32.4 24.2 [15] Inner Mongolia 33.3 28.6 22.2 38.6 59.0 34.7 41.0 24.1 [22] Jilin 31.3 25.5 24.8 25.2 22.0 18.8 23.9 23.7 [31] Shanghai 30.3 34.3 29.4 29.1 26.5 23.4 19.8 21.6 [43] Hunan 36.7 32.6 33.3 33.0 29.4 27.4 21.3 21.5 [46] Hainan 33.3 43.8 50.0 25.0 34.8 10.5 25.0 18.8 [53] Yunnan 15.5 17.5 29.4 32.4 30.3 22.7 19.2 18.4 [42] Hubei 20.0 32.3 21.4 16.7 15.5 15.5 15.3 18.2 [13] Hebei 23.1 27.1 22.2 21.1 22.1 20.0 19.7 17.6 [50] Sichuan⫹ Chongqing 38.8 35.7 46.8 36.7 39.8 42.0 34.3 17.5 [21] Liaoning 35.8 37.7 33.3 30.5 27.2 19.5 22.3 16.1 [34] Anhui 32.1 29.7 24.1 32.4 27.4 33.8 10.9 16.0 [62] Gansu 22.2 37.3 37.2 32.6 33.3 31.0 16.9 16.0 [44] Guangdong 21.4 25.9 27.3 30.0 24.4 19.3 16.9 14.3 [65] Xinjiang 15.8 19.1 18.9 20.3 26.3 21.3 18.5 13.9 [36] Jiangxi 34.5 38.2 40.3 38.5 36.8 21.3 20.0 13.6 [35] Fujian 14.9 18.2 20.0 20.0 9.7 12.1 25.2 13.3 [37] Shandong 17.9 16.3 14.8 18.1 16.6 13.3 9.7 9.2 [32] Jiangsu 17.3 18.2 19.1 17.8 18.0 13.8 10.7 8.0 [33] Zhejiang 15.8 17.5 15.7 23.6 15.6 10.8 10.0 5.9 Total 27.8 29.5 28.7 29.8 27.4 22.7 20.5 18.3
1 LINE SHORT REGULAR 1 LINE LONG 1995 1996 1997 1998 1999 2000 2001 2002 [21] Liaoning 89 87 95 117 77 46 94 114 [44] Guangdong 68 121 137 170 151 118 131 111 [50] Sichuan⫹ Chongqing 61 60 71 94 129 101 91 88 [32] Jiangsu 41 54 63 83 74 64 96 73 [31] Shanghai 34 50 63 82 72 47 63 69 [12] Tianjin 28 33 41 66 55 57 55 67 [37] Shandong 46 45 51 68 67 56 72 63 [22] Jilin 35 42 64 78 57 46 52 62 [42] Hubei 33 54 60 77 73 79 65 62 [23] Heilongjiang 47 59 58 104 69 74 74 56 [43] Hunan 24 31 45 54 37 48 41 55 [13] Hebei 28 30 40 54 47 48 64 53 [61] Shaanxi 31 34 38 49 53 46 48 45 [41] Henan 34 44 45 51 51 55 69 43 [62] Gansu 16 16 17 41 36 11 31 40 [35] Fujian 9 10 8 11 9 10 38 36 [36] Jiangxi 19 24 26 35 36 32 32 31
[54] Tibet⫹ Qinghai ⫹ Ningxia 14 17 21 26 21 25 10 30
[45] Guangxi 14 27 31 35 30 23 25 29 [14] Shanxi 15 20 24 54 53 38 35 28 [15] Inner Mongolia 18 16 16 30 36 24 30 28 [34] Anhui 21 22 25 52 49 25 33 26 [53] Yunnan 8 9 15 21 25 19 24 26 [65] Xinjiang 10 36 16 21 25 22 28 26 [11] Beijing 11 18 37 40 39 37 51 22 [33] Zhejiang 19 22 28 38 26 20 23 14 [52] Guizhou 15 13 20 18 16 11 11 11 [46] Hainan 4 8 11 8 12 4 6 4 Total 792 1,002 1,166 1,577 1,425 1,186 1,392 1,312
Table 16. Nonperforming debt ratios by region, estimated from reported profitability, 1995– 2002, by nonperforming debt ratio in 2002 (percent)
1995 1996 1997 1998 1999 2000 2001 2002
[62] Gansu 29.1 27.1 21.5 44.6 42.9 13.1 35.2 43.0
[12] Tianjin 25.9 30.0 34.5 42.6 35.5 39.6 35.9 38.5
[54] Tibet⫹ Qinghai ⫹ Ningxia 41.2 50.0 48.8 45.6 36.8 39.7 18.2 38.0
[43] Hunan 30.8 36.0 45.5 48.6 31.1 38.7 29.1 37.2 [22] Jilin 30.2 30.7 41.8 50.3 35.8 27.1 31.9 35.8 [36] Jiangxi 32.8 35.3 36.1 44.3 41.4 36.0 35.6 35.2 [65] Xinjiang 17.5 52.2 21.3 26.9 31.6 27.5 35.0 32.9 [45] Guangxi 23.7 43.5 44.3 43.2 35.3 25.3 27.8 32.2 [21] Liaoning 30.4 28.0 27.5 32.3 24.1 14.3 28.1 32.2 [15] Inner Mongolia 34.6 28.6 25.4 42.9 45.6 32.9 36.1 31.8 [61] Shaanxi 39.7 38.6 38.4 48.5 39.6 34.6 33.8 28.8 [23] Heilongjiang 31.3 39.6 34.9 54.7 37.7 38.7 38.9 28.6 [50] Sichuan⫹ Chongqing 33.3 30.2 30.5 39.0 40.4 34.2 30.0 27.1 [46] Hainan 36.4 53.3 61.1 40.0 50.0 20.0 37.5 25.0 [42] Hubei 24.4 32.9 33.0 35.5 33.5 34.1 28.5 24.5 [35] Fujian 19.1 18.2 14.8 20.4 14.5 15.2 28.1 24.0 [53] Yunnan 13.8 14.1 22.1 28.0 33.3 25.3 23.3 22.8 [14] Shanxi 19.0 22.7 24.0 41.9 41.4 27.9 27.8 21.2 [13] Hebei 19.6 18.1 21.7 27.1 22.1 21.4 25.8 20.6 [44] Guangdong 22.8 34.8 33.1 39.0 34.8 25.2 25.7 19.9 [31] Shanghai 14.7 18.2 19.5 24.4 20.7 14.1 18.1 18.4 [34] Anhui 25.9 24.2 23.1 48.1 43.0 19.2 25.8 17.9 [41] Henan 25.6 28.0 27.3 25.1 24.3 25.1 29.5 16.9 [32] Jiangsu 18.1 20.1 21.1 25.9 22.6 17.3 21.1 14.5 [52] Guizhou 34.1 32.5 41.7 30.5 25.0 14.9 15.3 14.3 [11] Beijing 11.2 16.1 28.2 24.1 25.3 23.4 27.3 13.0 [37] Shandong 17.6 15.3 15.2 19.2 16.1 12.8 15.8 12.9 [33] Zhejiang 15.7 15.4 18.3 24.2 16.4 12.0 13.5 7.5 Total 24.1 27.0 27.8 34.2 29.7 23.9 26.1 22.9
1 LINE SHORT REGULAR 1 LINE LONG T able 17. Regression summarizing trend and cross-ownership patterns of nonperforming debt ratios (dependent variable nonperforming debt ratio) Nonperforming debt ratio estimated from imputed profit-ability as shown in table 7 Nonperforming debt ratio esti-mated from reported profitability as shown in table 8 Independent variables Coef ficient t-statistic Coef ficient t-statistic Inter cept 2,996.8 ⫺ 10.9 1,092.6 ⫺ 2.0 Y ear ⫺ 1.5 ⫺ 10.7 ⫺ 0.5 ⫺ 1.9 Ownership ⫽ private enterprises ⫺ 21.3 ⫺ 18.2 ⫺ 13.9 ⫺ 5.9 Ownership ⫽ collective enterprises ⫺ 18.2 ⫺ 15.5 ⫺ 10.7 ⫺ 4.6 Ownership ⫽ mixed-ownership domestic enterprises ⫺ 18.3 ⫺ 15.7 ⫺ 16.3 ⫺ 7.0 Ownership ⫽ for eign-invested enterprises ⫺ 13.3 ⫺ 11.4 ⫺ 5.1 ⫺ 2.2 Ownership ⫽ enterprises with investment fr om Hong Kong, Macau, and T aiwan ⫺ 13.6 ⫺ 11.7 ⫺ 6.9 ⫺ 3.0 Ownership ⫽ whole sample ⫺ 12.4 ⫺ 10.6 ⫺ 11.0 ⫺ 4.7 Ownership ⫽ state-owned enterprises 0 0 Number of observations 56 56 Adjusted R 2 0.906 0.538 Note: Regr ession equation: NPD ratio f(year ,ownership). W eighted least-squar es regr ession with weights of 100 to 2002, 95 to 2001, 90 to 2000, 85 to 1999, 80 to 1998, 75 to 1997, 70 to 1996, and 65 to 1995.
in that year. The NPD ratio for the whole sample for a given year is likely to be 12.4 percent lower than the NPD ratio for SOEs in that year.
The regression results shown in tables 17–19 can be used to make rough predictions about NPD ratios for a particular group in the future. But these rough predictions are based only on the pattern of NPD ratios during 1995–2002. Figures 1 and 2 show
1 LINE SHORT REGULAR 1 LINE LONG Table 18. Regression summarizing trend and cross-industry patterns of nonperforming debt ratios
(depend-ent variable nonperforming debt ratio)
Nonperforming debt ratio based on imputed profit-ability as shown in table 9
Nonperforming debt ratio based on reported profit-ability as shown in table 11
Independent variables Coefficient t-statistic Coefficient t-statistic
Intercept 3,571.4 15.7 1,754.7 ⫺6.4 Year ⫺1.8 ⫺15.6 ⫺0.9 ⫺6.3 [07] Petroleum extract ⫺25.3 ⫺11.1 ⫺17.3 ⫺6.3 [44] Electric power ⫺13.9 ⫺6.1 ⫺13.1 ⫺4.8 [16] Tobacco ⫺26.6 ⫺11.7 ⫺13.0 ⫺4.7 [24] Cultural ⫺7.1 ⫺3.1 ⫺8.5 ⫺3.1 [27] Medical ⫺15.6 ⫺6.9 ⫺7.8 ⫺2.9 [23] Printing ⫺1.2 ⫺0.5 ⫺7.5 ⫺2.7 [18] Garments ⫺9.5 ⫺4.2 ⫺5.3 ⫺1.9 [21] Furniture ⫺7.3 ⫺3.2 ⫺4.6 ⫺1.7 [43] Other manufacturing ⫺8.2 ⫺3.6 ⫺4.0 ⫺1.5
Industry⫽ whole sample ⫺8.4 ⫺3.7 ⫺4.0 ⫺1.5
[26] Raw chemicals ⫺8.4 ⫺3.7 ⫺3.0 ⫺1.1
[10] Nonmetal mining ⫺3.2 ⫺1.4 ⫺2.1 ⫺0.8
[15] Beverage ⫺18.0 ⫺7.9 ⫺1.4 ⫺0.5
[40] Electric equipment ⫺5.4 ⫺2.4 ⫺1.0 ⫺0.4
[41] Electronic and telecommunications 1.0 0.4 ⫺0.6 ⫺0.2
[25] Petroleum processing ⫺15.5 ⫺6.8 ⫺0.5 ⫺0.2 [22] Papermaking ⫺11.3 ⫺5.0 ⫺0.1 ⫺0.0 [46] Tap water 0.0 0.0 [19] Leather ⫺5.9 ⫺2.6 0.0 ⫺0.0 [06] Coal mining 8.6 3.8 0.2 ⫺0.1 [33] Pressing nonferrous ⫺6.9 ⫺3.0 0.6 ⫺0.2 [09] Nonferrous mining 7.3 3.2 0.7 ⫺0.3 [08] Ferrous mining ⫺1.3 ⫺0.6 1.4 ⫺0.5 [29] Rubber ⫺7.2 ⫺3.2 1.8 ⫺0.6 [37] Transport equipment ⫺0.3 ⫺0.1 2.1 ⫺0.8 [35] Ordinary machinery 1.0 0.4 3.3 ⫺1.2 [30] Plastic ⫺4.9 ⫺2.1 3.8 ⫺1.4 [34] Metal products ⫺1.2 ⫺0.5 4.2 ⫺1.5 [32] Pressing ferrous ⫺4.9 ⫺2.1 4.4 ⫺1.6 [31] Nonmetal products ⫺6.8 ⫺3.0 4.6 ⫺1.7 [28] Chemical fiber ⫺3.5 ⫺1.5 4.6 ⫺1.7 [12] Timber logging 4.7 2.1 4.6 ⫺1.7 [14] Food production ⫺4.8 ⫺2.1 5.1 ⫺1.8 [17] Textile ⫺5.0 ⫺2.2 5.4 ⫺2.0 [42] Instruments 9.5 4.2 5.5 ⫺2.0 [36] Special equipment 6.9 3.0 6.9 ⫺2.5 [13] Food processing ⫺4.9 ⫺2.2 8.7 ⫺3.1 [20] Timber 2.6 1.1 8.7 ⫺3.2 [45] Gas production 35.0 15.4 22.1 ⫺8.0 Number of observations 312 312 Adjusted R2 0.850 0.610
Note: Regression equation: NPD ratio f(year, industry). Weighted least-squares regression with weights of 100 to 2002, 95 to 2001, 90 to 2000, 85 to 1999, 80 to 1998, 75 to 1997, 70 to 1996, and 65 to 1995. See appendix 2 for full industry names corresponding to in-dustry codes.
